We have performed experiments to improve our understanding of the molecular mechanisms of deregulated differentiation pathways in TSCs: The delicate balance between stem cell self-renewal and differentiation is controlled by various cell intrinsic and extrinsic factors that are critical for normal tissue homeostasis. Despite extensive phenotypic and functional similarities between GSCs and normal stem cells, the differentiation potentials of GSCs are not entirely normal. Elucidation of the differentiation pathways that are operative in both normal stem cells and GSCs will be critical for fully understanding tumorigenesis and will likely lead to novel therapeutic targets. We have also identified a set of deregulated differentiation pathways in GSCs derived from human primary GBM. Elucidation of underlying molecular mechanism will provide important clues for predicting sensitivity of differentiation therapeutic approach. Characterization of TSCs in aspect of differentiation-inducing agents further revealed the limitations of traditional glioma cell lines grown in serum. For example, retinoic acid treatment and CNTF exposure potently induce differentiation in most GBM tumor initiate cells (TICs) but not of traditional cell lines. This prompted us to question whether many of potential tumor suppressors and/or cytostatic genes previously studied in cell lines, were not recognized. Given the ever-increasing number of potential GSGs and oncogenes in glioblastoma TSCs identified from bioinformatics approaches and technical expertise of stem cell culture accumulated in the laboratories, we have set up screening systems to study the function of these genes in stem cell cultures. In addition, we have made significant progress on one of keystone projects that is to understand the genomic and molecular signaling similarities and differences between our glioma TSCs and normal neural stem cells (NSC). We have performed a very large scale study of 7 different GBM-derived TSCs and normal embryonic NSC lines under both proliferative and differentiating conditions and derived high-throughput mRNA and microRNAs profiling. Since November of 2014, when Dr Gilbert initiated the new glioma stem cell project, we have created 8 more glioma-derived TSCs, introduced 3 low-grade TSCs, and subsequently performed the computational analyses for characterization of the genetics and signaling pathways in these GSCs, enabling these cell lines to be used to explore the therapeutic potential of glioma stem cell biology. As described above, the evolving collaboration with the NCI Patient Derived Models Repository will increase our portfolio of tumor models over the coming year.